Vascular age estimation using a consumer wearable sleep tracker

Area of Science:

• Gerontology and cardiovascular research.
• Biomedical engineering and wearable technology.

Background:

• Traditional vascular aging assessment relies on costly clinical equipment.
• Wearable health trackers offer a scalable alternative for longitudinal vascular aging monitoring.

Purpose of the Study:

• To evaluate a consumer wearable (Oura Ring) for detecting age-related photoplethysmography (PPG) waveform differences.
• To compare the performance of a wearable PPG sensor against a clinical-grade device in estimating vascular age.

Main Methods:

• Simultaneous overnight PPG data collection from a fingertip pulse oximeter and an Oura Ring in 160 healthy adults.
• Extraction of pulse waveform features and estimation of vascular age using a deep learning model.
• Comparison of waveform feature correlations with age and prediction performance between devices.

Main Results:

• Age-related changes were most prominent in PPG crest time for both devices.
• The reflection index showed weaker age correlation for the ring sensor compared to the fingertip device.
• The deep learning model achieved comparable vascular age prediction performance (r ≈ 0.80-0.84) across both clinical-grade and consumer-grade devices.

Conclusions:

• Nocturnal PPG from wearable devices is feasible for assessing vascular age.
• Consumer wearables provide a scalable solution for longitudinal monitoring of vascular aging.